breeding potential and genetic diversity of “híbrido do ...brazilian society of plant breeding....

298 Crop Breeding and Applied Biotechnology 10: 298-304, 2010

TA Setotaw et al.

Crop Breeding and Applied Biotechnology 10: 298-304, 2010

Brazilian Society of Plant Breeding. Printed in Brazil

Received 10 December 2009

Accepted 20 June 2010

Breeding potential and genetic diversity of “Híbrido doBreeding potential and genetic diversity of “Híbrido doBreeding potential and genetic diversity of “Híbrido doBreeding potential and genetic diversity of “Híbrido doBreeding potential and genetic diversity of “Híbrido doTimor” coffee evaluated by molecular markersTimor” coffee evaluated by molecular markersTimor” coffee evaluated by molecular markersTimor” coffee evaluated by molecular markersTimor” coffee evaluated by molecular markers

Tesfahun Alemu Setotaw1, Eveline Teixeira Caixeta 2*, Guilherme Ferreira Pena1, Eunize Maciel Zambolim1, Antonio Alves Pereira3

and Ney Sussumu Sakiyama1

ABSTRACT - AFLP, RAPD and SSR molecular markers were used to study the genetic diversity and genetic structure of theHíbrido de Timor germplasm. The principal coordinate analysis, UPGMA cluster analysis based on genetic dissimilarity ofJaccard, Bayesian model-based cluster analysis, percentage of polymorphic loci, Shannon’s information index and Nei genediversity were employed to assess the genetic diversity. The analyses demonstrated a high genetic diversity among Híbrido deTimor accessions. UPGMA and Bayesian cluster analyses grouped the accessions into three clusters. The genetic structure ofHíbrido de Timor is reported. The management of Híbrido de Timor germplasm variability and its potential use in breedingprograms are discussed.

Key words: Coffee breeding, DNA markers, rust resistance, Bayesian analysis, principal coordinate analysis.

INTRODUCTION

“Híbrido de Timor” coffee, a natural interspecifichybrid between C. arabica L and C. canephora Pierre exFroehner was first found in a plantation of cultivar Typicain Timor Island in 1917 (Bettencourt 1973). The derivativesof this coffee tree have been used as a source of resistancegene for economically important diseases and pests ofcoffee such as coffee leaf rust (Hemileia vastatrix), coffeeberry disease (CBD) caused by Colletotrichum Kahawae,root knot nematode (Meloidogyne exigua) and bacteriosiscaused by Pseudomonas syringae pv garçae (Bettencourt1973, Chaves 1976, Goncalves and Pereira 1998, Carvalhoet al. 1989, Pereira et al. 2005). Cultivars derived from“Híbrido de Timor” were released for production in Kenya,Brazil, Colombia and Costa Rica (Charrier and Eskes 1997,

Lashermes et al. 2000, Sera 2001, Pereira et al. 2005, Sera etal. 2008).

The first introduction of “Híbrido de Timor”accessions to Brazil date back from 1976 via vegetativepropagation and seeds from CIFC (Centro da Investigaçãode Ferrugem do Café, Portugal), IIAA (InstitutoInvestigação Agronomia de Angola) and ERU (EstaçãoRegional de Uige) (Pereira et al. 2002). These materialswere used extensively in the breeding program of coffeefor resistance to diseases and pests in Brazil.

Since “Híbrido de Timor” is an important gene sourcefor resistance to diseases and used to a large extent in thebreeding program of coffee in the world, understandingthe existing genetic variability will be very important andit will help in designing the breeding strategies to developcultivars which combine good cup quality with resistance

1 Universidade Federal de Viçosa (UFV), Laboratório de Biotecnologia do Cafeeiro (BIOCAFÉ), BIOAGRO, 36570-000, Viçosa, MG, Brazil.2 Embrapa Café, 70770-901, Brasília, DF, Brazil. Current address: UFV, BIOCAFÉ. *E-mail: [email protected] Empresa de Pesquisa Agropecuária de Minas Gerais, Centro Técnico da Zona da Mata Mineira, Vila Gianeti, casas 46 e 47, C.P. 216, 36570-000, Viçosa, MG,

Brazil.

Crop Breeding and Applied Biotechnology 10: 298-304, 2010 299

Breeding potential and genetic diversity of “Híbrido do Timor” coffee evaluated by molecular markers

to disease. In addition, it is important to plan appropriateconservation strategies to maintain and conserve theaccessions of “Híbrido de Timor”. The study of geneticdiversity among accessions also helps to identify theduplicated accessions in the germplasm bank and planningappropriate selections to form the core collection. Theexistence of considerable genetic diversity within theHíbrido de Timor germplasm was reported by Lashermeset al. (2000) but he included only two accessions. So it isimportant to study the genetic variability including moreaccessions of “Híbrido de Timor”. To study the geneticvariability, morphological, isoenzyme and moleculartechniques can be used. Among them, the DNA moleculartechnique is the best and able to detect differences at thegenome level. Among the molecular technique SimpleSequence Repeats (SSR), Random Amplified PolymorphicDNA (RAPD) and Amplified Fragment Length Polymorphism(AFLP) are more frequently used for genetic diversity studyand proved to be efficient in establishing the corecollection. Teixeira-Cabral et al. (2002) and Teixeira-Cabralet al. (2004) used RAPD molecular markers to study thegenetic diversity and characterize the coffee differentialsof Hemileia vastatrix races. Lashermes et al. (2000), Orozco-Castillo et al. (1994), Silvestrini et al. (2007) and Silvestriniet al. (2008) used AFLP, RAPD and SSR techniques to

study the genetic diversity of coffee species. Maluf et al.(2005) used RAPD, AFLP and SSR molecular markers tostudy the genetic diversity of arabica coffee and detectedsignificant polymorphism among lines. EST-SSR molecularmarkers were also used to study the genetic diversity ofcoffee trees including accessions of “Híbrido de Timor”(Missio et al. 2009).

This work was realized with the objective ofinvestigating the existing genetic diversity among theaccessions of “Híbrido de Timor” using RAPD, AFLP andSSR molecular markers.

MATERIAL AND METHODS

Genetic materials and DNA extraction

Forty eight accessions from the germplasm bank ofUFV (Universidade Federal de Viçosa)/EPAMIG (Empresade Pesquisa Agropecuária de Minas Gerais) were analyzed(Table 1).

The DNA of the genotypes was extracted accordingto the method described by Diniz et al. (2005) from younggreen leaves collected from each genotype. The DNAconcentration was quantified using SpectrophotometerSmart Spec of BioRad. The extracted DNA was diluted

Table 1. Coffee accessions of Híbrido de Timor germplasm investigated for genetic diversityin this research

a Clones introduced from CIFC, Portugal; b Seeds introduced from IIAA, Angola, derived fromseeds of plants selected in CIFC; c Seeds introduced from CIFC, Portugal; d Seeds introducedfrom ERU, Angola, derived from seeds of plants selected in CIFC.


TA Setotaw et al.

Data analysis

The gels of RAPD, SSR and AFLP were scored forthe presence and absence of clearly amplified fragments.The data matrix was coded according to Falush et al. (2007)to fit for the analysis of Structure program assuming acorrect model of dominance. AFLP data statistical package(Ehrich 2006) was used to manage the data conversionfrom txt format to Structure and Arlequin data format.

To study the genetic diversity, the Jaccard geneticdistance (1- Jaccard similarity coefficient) (Jaccard 1908),Nei genetic diversity index (Nei 1973), Shannon´s InformationIndex and percent polymorphic bands (P %) were estimated,using Popgene software version 1.3 (Yeh and Boyle 1997).Principal coordinate analysis (PCoA) was performed usingthe GenAlex 6.2 population genetic analysis software(Peakall and Smouse 2006).

To define the population structure of the “Híbridode Timor” coffee accessions, the Bayesian model-basedclustering method of Pritchard et al. (2000) was implementedin the Structure 2.1 software (http://www.pritch.bsd.uchicago.edu/) to estimate population admixture throughinferred ancestry. For each of the K=2 to K=12 settings, 20independent simulations were performed using theadmixture model and 5000 replicates for burn-in and postburning sampling by Markov Chain Monte Carlo of 50,000runs to estimate the number of subpopulations for each ofthe k values. The appropriate number of clusters (K) wasdetermined according to Evanno et al. (2005).

RESULTS AND DISCUSSION

Genetic diversity analysis

The AFLP, RAPD and SSR molecular markersproduced 80, 120 and 25 polymorphic alleles amongaccessions of “Híbrido de Timor”, respectively. Thepopulation diversity parameters estimated on “Híbrido deTimor” coffee was presented on Table 2. These geneticdiversity measures showed that the Híbrido de Timorgermplasm has considerable genetic diversity.

The estimated Genetic dissimilarity coefficient (1-Jaccard similarity coefficient) was used in principalcoordinate analysis (PCoA). Before doing the PCoA, theMantel test between the distance matrix produced byAFLP, RAPD and SSR variation was performed. The testshowed significant correlation among them so that thecombined data was used for further analysis. PCoA usingthe combined data from AFLP, RAPD and SSR markers

with TE (Tri-HCL 10mM, EDTA 1mM, pH 8.0) to aconcentration of 10 ηg μL-1 for RAPD, 25 ηg μL-1 for SSRand 50 ηg μL-1 for AFLP analysis.

RAPD molecular marker

A total of 52 RAPD oligonucleotide primers (OperonTechnologies) were analyzed. Each reaction constituted atotal volume of 25 Ml, with 2.5 μL of 10 ηg μL-1 genomicDNA, 1.25 μL of Taq DNA polymerase, 0.25 μL of eachdNTP, 1 μL of primer, 2.5 μL of KCl, 2.5 μL of Tris HCl pH8.3 and 2 mM of MgCl2. The PCR reaction was done usinga thermo-cycle Gene Amp PCR System 9600. For theamplification, the DNA was denatured at 95 oC for 1 minutefollowed by 40 cycles at 94 oC for 15 second to denatureDNA, 35 oC for 30 seconds to anneal the primer, 72 oC for1 minute extension of the primer followed by final extensionto complete the process of amplification at a temperatureof 72 oC for 7 minutes and stored at 4 oC until used. Theamplified DNA was separated using agarose gelelectrophoreses 1.2 %. The gel was then immersed in thesolution of Etidium Bromide and visualized using the EagleEye II Still Video System.

SSR marker

Eighteen microsatellite primers obtained fromCombes et al. (2000) and Rovelli et al. (2000) were analyzed.The PCR reaction was realized in a total volume of 20 μLcontaining 50 ηg of DNA, 0.6 unit of Taq DNA polymerase,buffer 1x, 1mM of MgCl2, 150 μM of each dNTP and 0.1 μMof each primer. The amplification was done using theprocedure of touchdown-PCR that consisted of denaturationat 94 °C for 2 minutes, followed by 13 cycles of denaturationat 94 °C for 30 seconds, primer annealing at 67 °C to 55 °Cfor 30 seconds, reducing 1i°C of each cycle and anextension of primer at 72 °C for 30 seconds. This step wasfollowed by 30 more cycles of denaturation at 94 °C for 30seconds, primer annealing at 55 °C for 30 seconds andprimer extension at 72 °C for 30 seconds. The finalextension was done at 72 °C for 8 minutes. The PCRproducts were separated on a 6 % denaturing polyacrylamidegel and visualized by a silver staining solution.

AFLP analysis

The AFLP genotyping of coffee accessions was doneaccording to the method described by Brito et al. (2010). Theprimer combinations MseI-AGC with EcorI-CGT and MseI-AGC with EcorI-CTC were used to genotype the coffeeaccessions in this study.



formed a clear grouping pattern (Figure 1). The first twoprincipal coordinates explained 66.83% of the totalvariation among the accessions (Figure 1). In addition,the clustering analysis using the UPGMA method basedon the combined data of the three molecular markersformed three principal cluster groups (Figure 2). TheCIFC4106, considered the first plant found in the TimorIsland (Pereira et al. 2005), grouped together with CIFC832/1, CIFC 832/2, and CIFC 1343/269 introduced in CIFCfrom Timor Island by seed (Betencourt 1973) and sent toBrazil via vegetative propagation. These groups alsoincluded accessions of “Híbrido de Timor” introduced byseed to UFV from CIFC.

Both the genetic distance based analysis of diversityand population based statistics (percent polymorphic loci,Nei gene diversity index and Shannon´s index) demonstratedthe efficiency of the three molecular markers in detectinggenetic diversity among “Híbrido de Timor” accessionsfound in the germplasm bank of UFV/Epamig.

Genetic Structure

To understand the genetic structure of the “Híbridode Timor” accessions the clustering analysis based onthe Bayesian statistical analysis was performed using theStructure program (Pritchard et al. 2000). This analysis,using the data from the three molecular markers, groupedthe accessions into three clusters. Eighty nine percent ofthe accessions were assigned in their respective groupwith a shared ancestral probability higher than 0.81 (Table 3).CIFC 4106, considered the first “Híbrido de Timor” plantfound on Timor Island (Pereira et al. 2005), grouped withthe other CIFC materials with a shared ancestral probabilitygreater than 90 %. This result agreed with the historicalinformation that this plant gave way to the origin of these“Híbrido de Timor” accessions. Under Viçosa soil andclimatic condition, CIFC 4106 had high flowering capacitybut produced a small amount of fruit, all with moka typeseeds, which indicates the occurrence of self-incompatibility(Pereira et al. 2008). Additional field observations showedthat no artificial hybridization succeeded, when CIFC 4106was used as the female parent. On the other hand, when

Table 2. The Nei gene diversity measure, Shannon´s informationindex and percent polymorphic bands (P %) obtained using AFLP,RAPD and SSR molecular markers for “Híbrido de Timor” coffeeaccessions

’

Figure 2. The dendrogram produced based on Genetic dissimilarityof Jaccard estimated based on the combined data of AFLP, RAPDand SSR molecular markers using the UPGMA cluster method for“Híbrido de Timor” coffee

Figure 1. Principal coordinate analysis of the combined datafrom AFLP, RAPD and SSR diversity among accessions of “Híbri-do de Timor” coffee (Numbers correspond to the code of theaccessions in Table 1).


TA Setotaw et al.

Table 3. Shared ancestral probability of each accessions of “Híbrido de Timor” in relation to three clusters inferred basedon AFLP, RAPD and SSR variation using Bayesian clustering analysis



used as a male parent (donor of pollen), there was somesuccess in crossing with a low percentage of fertilization(Pereira et al. 2008). These field observations support thehypothesis that CIFC 4106 originated from naturalinterspecific hybridization between C. arabica andC.icanephora, as suggested by Pereira et al. (2005).

The combined data from RAPD, SSR and AFLPmolecular markers grouped the accessions CIFC 832/1,CIFC 832/2 and CIFC 1343/269 in the same cluster with ashared ancestral probability greater than 68.9%. Theseaccessions were the first genetic material, collected fromthe same plant on Timor Island, sent to CIFC (Bettencourt1973). These plants were introduced to Brazil throughvegetative propagation. Still these accessions were usedfor the development of arabic coffee cultivars with durableand complete resistance to coffee leaf rust all over theworld (Bettencourt 1973, Pereira et al. 2005, Zambolim etal. 2005).

All the accessions derived from UFV 377 and fiveaccessions derived from UFV 376 (UFV 376-01, UFV 376-05, UFV 376-52, UFV 376-57 and UFV 376-79) were groupedtogether in Cluster II with a shared ancestral probabilitygreater than 0.81. The other three accessions derived fromUFV 376 shared some level of allele with group II and III.These accessions have a common origin. A plant designatedas VCE1587, resistant to leaf rust, was selected in Tanzania.Seeds of this plant were taken to CIFC, from which twoplants were selected and designated as CIFC 2234 and2235. Later on, seeds from CIFC 2234 and 2235 were sentto IIAA (Instituto de Investigação Agronômica de Angola),under the code IIAA808 and IIAA811, respectively. TheUFV376 and UFV377 originated from seeds collected inthe IIAA808-5 and IIAA811-5 plants, respectively.

UFV 437, UFV 438, UFV 439, UFV 440, UFV 441, UFV442, UFV 443, UFV 445, UFV 446, UFV 447 and UFV 448were grouped in Cluster I. These accessions were derivedfrom CIFC 2570 (Table 1). The accessions that originatedfrom ERU202 (derived from CIFC 1343/136) were groupedin Cluster III. This cluster was also composed of coffee

derived from CIFC 2252, CIFC 1590/9, CIFC2568 andCIFC2571.

In general the grouping pattern using the Bayesianmodel-based clustering method (Pritchard et al. 2000) wasin agreement with the origin of accessions. Most of theaccessions derived from the same plant by seedpropagation grouped together with a shared ancestralprobability > 0.80. López-Gartner et al. (2009) also provedthe efficiency of the model based Bayesian clustering analysisfor studying the genetic diversity and structure of arabiccoffee originating from Ethiopia.

The economic and environmental benefit and thehigh cup quality aspects of the arabica coffee cultivarsderived from “Híbrido de Timor” were previously reported(Pereira et al. 2005, Pereira et al. 2008, van de Vossen 2009).Our data clearly showed that the genetic variability of theHíbrido de Timor germplasm is potentially useful for breedingprograms, especially because of the well known narrowgenetic basis of C. arabica. Additionally, we have organizedthe variability of this germplasm, in such a way that a breedercan choose parental plants according to the objectives inhis/her breeding program. For instance, a new parentalplant can be chosen based on the criteria of the geneticsimilarity or dissimilarity with CIFC 832/1 and CIFC 832/2,the major sources of genes for resistance to coffee leaf rust(Bettencourt 1973), depending on the disease resistance,productivity, cup quality, and other characteristicsconsidered. Our results can also be applied to germplasmmanagement. A small number of each of the three clustersidentified in this study can well represent the Híbrido deTimor germplasm if a sub-collection is needed, with theadvantage of preserving the original variability with lowcost and labor.

ACKNOWLEDGEMENTS

The authors wish to thank CNPq and TWAS for thefinancial support and scholarships.


TA Setotaw et al.

REFERENCESBettencourt A (1973) Considerações gerais sobre o ‘Híbrido

de Timor’. Secretaria da Agricultura do Estado de São Paulo,Campinas, 20p. (Circular 23).

Brito GG, Caixeta ET, Gallina AP, Maciel-Zambolim E, ZambolimL, Diola V and Loureiro ME (2010) Inheritance of coffee leafrust resistance and identification of AFLP markers linked tothe resistance gene. Euphytica 173: 255-264.

Carvalho A, Fazuoli LC and Costa WM (1989) Produtividade doHíbrido de Timor, de seus derivados e outras fontes de resistênciaa Hemileia vastatrix. Bragantia 48: 73-86.

Charrier A and Eskes A (1997) Coffee. In Charrier A, Jacquot S,Hamon D and Nicolas C (eds.) Tropical plant breeding. Cirad,Repères, p. 128-152.

Chaves GM (1976) Melhoramento do cafeeiro visando à obtençãode cultivares resistentes à Hemileia vastatrix Berk et Br. RevistaCeres 23: 321-332.

Combes MC, Andrzejewski S, Anthony F, Bertrand B, Rovelli P,Graziosi G and Lashermes P (2000) Characterization ofmicrosatellites loci in Coffea arabica and related coffee species.Molecular Ecology 9: 1171-1193.

Diniz LEC, Sakiyama NS, Lashermes P, Caixeta ET, Oliveira ACB,Maciel-Zambolim E, Loureiro ME, Pereira AA and ZambolimL (2005) Analysis of AFLP marker associated to the Mex-1resistance locus in Icatu progenies. Crop Breeding andApplied Biotechnology 5: 387-393.

Potencial de melhoramento e diversidade genética doPotencial de melhoramento e diversidade genética doPotencial de melhoramento e diversidade genética doPotencial de melhoramento e diversidade genética doPotencial de melhoramento e diversidade genética docafé “Híbrido de Timor” avaliado por marcadorescafé “Híbrido de Timor” avaliado por marcadorescafé “Híbrido de Timor” avaliado por marcadorescafé “Híbrido de Timor” avaliado por marcadorescafé “Híbrido de Timor” avaliado por marcadoresmolecularesmolecularesmolecularesmolecularesmoleculares

RESUMO - Marcadores moleculares AFLP, RAPD e SSR foram usados para estudo da diversidade e estrutura genética dogermoplasma de Híbrido de Timor. As análises de coordenadas principais, de agrupamento UPGMA baseado na dissimilaridadegenética de Jaccard, de agrupamento baseado no modelo Bayesiano, a percentagem de locos polimórficos, o índice deShannon e a diversidade genética de Nei foram utilizadas para acessar a diversidade genética. As análises demonstraramalta diversidade genética entre os acessos de Híbrido de Timor. A análise de agrupamento UPGMA e o modelo Bayesianopermitiram a alocação dos acessos em três grupos. A estrutura genética dos acessos de Híbrido de Timor foi descrita. Amanutenção e manuseio da variabilidade do germoplasma de Híbrido de Timor e seu potencial uso nos programas demelhoramento de cafeeiro foram discutidos.

Palavras-chave: Melhoramento do cafeeiro, marcadores de DNA, resistência à ferrugem, análise bayesiana, análise de coordenadasprincipais.

Ehrich D (2006) AFLPdat: a collection of R functions forconvenient handling of AFLP data. Molecular Ecology Notes6: 603-604.

Evanno G, Regnaut S and Goudet J (2005) Detecting the number ofclusters of individuals using the software Structure: A simulationstudy. Molecular Ecology 14: 2611-2620.

Falush D, Stephens M and Pritchard JK (2007) Inference ofpopulation structure using multilocus genotype data: dominantmarkers and null alleles. Molecular Ecology Notes 7: 574-578.

Gonçalves W and Pereira AA (1998) Resistência de cafeeiros anematóides IV-reação de cafeeiros derivados de Híbridos deTimor a Meloidogyne exigua. Nematologia Brasileira 22:39-50.

Jaccard P (1908) Nouvelles recherches sur la distribution florale.Bulletin de la Societé Vanddoise des Sciences Natureles44: 223-270.

Lashermes P, Andrzejewski S, Bertrand B, Combes MC, Dussert S,Graziosi G, Trouslot P and Anthony F (2000) Molecular analysisof introgressive breeding in coffee (Coffea arabica L.).Theoretical and Applied Genetics 100: 139-146.

López-Gartner G, Cortina H, McCouch SR and Moncada MDP(2009) Analysis of genetic structure in a sample of coffee(Coffea arabica L.) using fluorescent SSR markers. TreeGenetics and Genomes 5: 435-446.

breeding potential and genetic diversity of “híbrido do ...brazilian society of plant breeding....

Documents